GamjaPower/baseline-prepare-dataset

该数据集包含用于自然语言处理任务的训练和验证数据，特征包括input_ids、token_type_ids、attention_mask和next_sentence_label。数据集分为训练集和验证集，训练集包含2,353,324个样本，验证集包含260,928个样本。总下载大小为779,432,338字节，数据集总大小为8,083,267,184字节。

This dataset contains training and validation data for natural language processing tasks, with features including input_ids, token_type_ids, attention_mask, and next_sentence_label. The dataset is divided into a training set with 2,353,324 examples and a validation set with 260,928 examples. The total download size is 779,432,338 bytes, and the total dataset size is 8,083,267,184 bytes.